Kinetics, mechanism, and isotope rate effect of the base-catalyzed

Nucleophilic substitution at acetylenic carbon. The last holdout. Sidney I. Miller and Jerome I. Dickstein. Accounts of Chemical Research 1976 9 (10),...
2 downloads 0 Views 1MB Size
1602

Kinetics, Mechanism, and Isotope Rate Effect of the Base-Catalyzed Formation of 1-Haloalkynes from Acetylenes and Sodium Hypohalites’s2 Ruey-Rong Lii and Sidney I. Miller*

Contribution f r o m the Department of Chemistry, Illinois Institute of Technology, Chicago, Illinois 60616. Received June 14, 1972 Abstract: The formation of I-haloalkynes in water at pH > I 1 and [OH-] > 2[X1] follows third-order kinetics, of = kr[RC==CH][OH-][OX-j. The solvent isotope rate effects for the chlorination at 46.1 O and bromination at 27.5 O of phenylacetylene are essentially absent, i.e., k C H H 2 0 / k C D D 2 0 = 1.01 and 1.07 (+0.04), respectively. For chlorination of phenylacetylene at 25.3“,k = 9.5 x M - 2 sec-l, AH* = 13.0 kcal/mol, and A S * = -27 eu. For bromination of four para-substituted phenylacetylenes (H, C1, Br, and CH,), AH* = 7.5-8.7 kcal/mol and - A S 7 =26-28 eu; Hammett pvalues forthis group were 0.72,0.77, and 0.825 at 35.3,25.3, and 15.3”,respectively, and the rate constants ( M - 2 sec-1) depend on temperature and Hammett u as follows: log k = -1.135~ 1893/T 574u/T 7.4. Positive and specific salt effects on the rate of chlorination and on the activity coefficient of phenylacetylene were noted. Toward phenylacetylene, the reactivity of hypohalite is in the order IO- > BrO- >> CIO-. Rates of chlorination and bromination of an alkylacetylene, 3-methylbut-I-yn-3-01, are within a factor of

+

+

5-10 of those for phenylacetylene. It is proposed that the mechanism of halogenation involves a fast preequilibrium to form RCrC-, a rate-determining step of this anion and hypohalite to yield the intermediate RC=CX02-, and a rapid reaction of this intermediate with water to produce I-haloalkyne and two hydroxide ions. One consequence of this mechanism is that the reverse process, I-dehalogenation, probably has the form L;,= k,[RC=CX]* [OH-I2, or its equivalent. In exploring the dehalogenation process, we found a rather different route with arsenious oxide; that is, the rate law is D = k[CsH,C~Br][As,03][OH-]2. The general significance of this work in the area of positive halogen chemistry is indicated.

T

he base-catalyzed halogenation of weak carbon acids, e.g., ketones, nitriles, sulfones, nitroalkanes, etc., has been extensively studied. This process is also of considerable synthetic importance for terminal acety l e n e ~but ~ has been used only infrequently for hydrocarbons-the preparations of polychloro- and polybromocyclopentadienes or indenes seem to be unique. 3b However, there seems t o be no precedent for rate studies of base-catalyzed halogenation of hydrocarbons. In this paper we demonstrate that this is possible with acetylenes, RC-CH

+ OX-

RC-CX

+ OH-

(1)

The mechanism of reaction 1 is unusual, first in that the halogenation is kinetically of the third order and second in that hypohalite ion is the attacking species.

u = k[RCrCH][OH-][OX-] (2) Among the hundreds of base-catalyzed halogenations ~ t u d i e d ,there ~ are perhaps two almost forgotten examples (hypochlorite with acetone or acetophenone) with the same rate law.5 The rediscovery and investiga(1) (a) This work, supported in part by Public Health Service Grant GM 7021, was abstracted from R.-R. Lii’s Ph.D. thesis, Illinois Institute of Technology, May 1971 ; (b) presented in part at the 158th National Meeting of the American Chemical Society, New York, N. Y., Sept 8-12, 1969, Paper 213. (2) Preliminary communication: R.-R. Lii and S. I . Miller, J . Amer. Chem. SOC.,91, 7524 (1969). All of the rate constants in this paper should be doubled; the factor 2 was inadvertently omitted in the analysis for free halogen. (3) (a) S. Y . Delavnrenne and H. G. Viehe in “The Chemistry of Acetylenes,” H. G. Viehe, Ed., Marcel Dekker, New York, N. Y., 1969, Chapter 10; (b) F. Straus, L. Kollek, and W. Heyn, Chem. Ber., 63, 1868 (1930); (c) S. I. Miller, G. R. Ziegler, and R. Wjelesek, Org. Syn., 45, 86 (1965). (4) (a) R. P. Bell, “The Proton in Chemistry,” Cornell University Press, Ithaca, N. Y . , 1959, Chapters X, XI; (b) R. P. Bell and B. G. Cox, J . Chem. SOC.B, 652 (1971); (c) F. Hibberr, F. A . Long, and E. A. Walters, J . Amer. Chem. Soc., 93, 2829 (1971). (5) P. D. Bartlett and J . R. Vincent, ibid., 57, 1596 (1935).

Journal of the American Chemical Society

95:5

tion of this relatively obscure mechanism has uncovered problems and provided insights in several important areas. We investigated the reaction of hypochlorite or hypobromite ions with diisopropyl ketone and found that rate law 2 applied a t high P H . ~ As a result, conclusions drawn by previous workers on mechanism, isotope effects, and tunneling effects in this system will probably have to be revised.‘j We examined the reaction of hypochlorite with nitroethane and found a novel rate law in which there was an inverse dependence in hydroxide ion at pH >11.7 In this paper, the focus turns to several aspects of positive halogen Chemistry. Appropriately, recent reports on apparently unrelated reactions also involve hypohalite and carbanions analogous to those in both forward and reverse reactions of eq 1, e.g., the complex conversion of picric acid to bromopicrin with a rate law similar to that of eq 28 and several base-catalyzed halogen abstractions from polyhaloaromatics, haloalkynes, propargyl halides, etc.gs10We shall touch on some of these aspects of positive halogen chemistry later. It will be recalled that in the description of basecatalyzed halogenation, say of a ketone, one usually sees the following ~ c h e m e . ~ , ~ a

CH+BeC-+BH+ -a

(3) 1)

C-+X++CX (6) R.-R. Lii and S. I . Miller, J . Chem. SOC.B, 2269 (1971). (7) R.-R. Lii and S. I. Miller, ibid., 2271 (1971). (8) A. R. Butler and H. F. Wallace, ibid., 1758 (1970). (9) Numerous citations to positive halogen abstraction are given by J . F. Bunnett, Accourlts Chem. Res., 5, 139 (1972). (10) (a) A . Fujii, I. I. Dickstein, and S. I. Miller, Tetrahedron Letr., 3435 (1970); (b) A. Fujii and S. I . Miller, J . Amer. Chem. SOC.,93,3694 (1971).

March 7, 1973

1603

If the steady-state assumption applies to the intermediate C-, one obtains the general rate law 4. In most cases, kPa[BH+] > kh[X+]. Bell and Long did this simply by having [Xi] -t 0 and halogenation proceed via molecular bromine or tribromide ion.4 In our work, we did this by suppressing the concentrations of all halogenating species but the least active one, namely, hypohalite ion. Admittedly, there are rate laws other than (4) and schemes other than (3) that may apply to a given halogenation; the steady-state condition, for example, need not hold and the halogenating agent may prefer to attack an isomeric form of CH, etc. Here we consider the alternatives in eq 3 and 4 with special reference to acetylenes. Experimental Section The following instruments were used routinely: Beckman IR-8 for infrared (ir) spectra; Varian A-60 for nuclear magnetic resonance (nmr) spectra; Varian ,MAT CH7 for mass spectra; Aerograph Models 20 and 700 for gas chromatography. Spectral checks were made on neat compounds, unless otherwise indicated; tetramethylsilane was used as an internal reference for nmr. All melting points and boiling points listed are uncorrected. Our acetylenes were prepared by established methods and most of them have been described recently.IO Here we add 4-bromophenylacetylene, mp 64-65' (lit.Ii 63-63.5"), ir P (CCla) 2090 cm-', iimr 6 (CCI,) 2.99 (1 H, s); mercuric 4-bromophenylacetylide, mp 255-2562 (lit.'? 256-256.5"). v(CCI4) 2130 cm-'; 4-bromophenylbromoacetylene. mp 96-97" (lit.Ir mp 93-95"), P (CCl4) 2180 cm-'; 4-nitrophenylacetylene, mp 149" (lit.11 148-149"). Phenylacetylene-d was prepared from sodium phenylacetylide and deuterium oxide in ether; nmr analysis indicated -90 atom deuteration. oc,P-Dichlor~styrene.~~ Phosphorus pentachloride (10.4 g) and w-chloroacetophenone (7.7 g) were warmed at ca. 100" until a clear yellow melt was obtained. Distillation at 1 atm yielded a fraction which had bp 210"; ir P (neat) 3070 cm-'; nmr 6 6.45 (1 H , s, trans), 6.63 (1 H, s, cis); uv A n l n r H 2 0 256 nm; mass spectrum m/e 172 (P)+, The cis and trans designations are 137 ( P - 35)', 102 (P - 70)'. tentative but are based on a fairly reliable chemical shift relationship for alkenes of this type.'& Hypohalites. Reagent grade chlorine, bromine. and iodine (resublimed) were used. Stock solutions of hypochlorite were prepared by distilling an appropriate amount (1 or 2 ml) of liquefied chlorine into a cooled (-0') solution of sodium hydroxide (250 ml. 4-8 M ) . Hypobromite or hypoiodite solutions were prepared by adding the halogen to a cooled (-0') solution of sodium hydroxide (20 ml. 0.45-0.9 M ) . These solutions were diluted with distilled water and used immediately, except for the sodium hypochlorite which was stored in a refrigerator. The kinetics of decomposition of our hypohalites have been i n ~ e s t i g a t e d , ~ ~but - l $ we also checked two of them briefly. Hypochlorite (A,, 292 mp) and hypobromite (A, 330 mp) were monitored spectrophotometrically. For the hypochlorite, there was no (11) A. D. Allen and C. D. Cook, Can. J . Chem., 41, 1084 (1963). (12) 7. H. Vaughn and J. A. Nieuwland, J . Amer. Chem. Soc., 56, 1207 (1934). (13) M. C. Verploegh, L. Donk, H. J. T. Bos, and W. Drenth, Recl. Trac. Chim. Pays-Bas, 90, 765 (1971). (14) R. Dyckeroff, Chem. Ber., 10, 119 (1877). (15) R. Tanaka, M. Rodgers, R. Simonaitis, and S. I. Miller, Tetrahedron, 27, 2651 (1971). (16) M. W. Lister, Can.J. Chem., 34,465 (1956). (17) B. Perlmutter-Hayman and G. Stein, J . Phys. Chem., 63, 734 (1959). (18) C. H . Li and C. F. White, J . Amer. Chem. SOC.,65,335 (1943). (19), (a) F . A . Cotton and G. Wilkinson, "Advanced Inorganic Chemistry, 2nd ed, Wiley, New York, N. Y . , 1966, p 570; (b) C. S . G. Phillips and R. J. P. Williams, "Inorganic Chemistry," Vol. I, Oxford University Press, London, 1965, Chapter 12.

significant decomposition within 1 week; for the hypobromite, decomposition was detectable in 2 days. Judging from the reported rate law for the decomposition of hypoiodide,l8its half-life is ca. four to eight times longer than that we found for the iodination of phenylacetylene at p K > 11 ; in any case, we found no trends in our kinetic data. la Sodium Deuterioxide. Fifty milliliters of 99.8 (Diaprep Inc.) deuterium oxide was placed in a 500 ml three-neck round-bottom flask equipped with a reflux condenser connected with a calcium chloride tube and a nitrogen inlet. While the flask was chilled in an ice bath and swept with nitrogen. an appropriate amount of small pieces of sodium metal, which had been freshly cut and washed with pentane, was added. The concentration of base was determined with standard hydrochloric acid. Analyses. A Corning 10 pH meter, equipped with a glass combination electrode, was used for acid-base titrations. The sodium hydroxide solutions were determined with standard (0.1000 M ) hydrochloric acid. Hypohalites were estimated iodimetrically : a known volume of hypohalite was treated with an excess of standard sodium arsenite, stirred, appropriately buffered, and backtitrated to the starch end point with standard iodine solution. 2u These titrimetric results were occasionally checked spectrophotometrically. In order to avoid the interference and presence of halate (X03-); the hypohalites were used and analyzed before their decomposition became significant. Kinetics. Process 1 was followed by uv spectroscopy in quartz cells (0.5. 1.0. or 5.0 cm path length). Beckman DK-2 and Cary 11 recording spectrophotometers equipped with thermostated cell compartments were used. Since the temperature within the cell was often slightly different (h0.5') from that of the external constant-temperature bath which was used to control the temperature of the cell compartment. both were checked. The stock solutions were left in the external bath ( > l o min) and then suitable volumes were mixed in the cell. Then the cell was quickly transferred to the cell compartment and absorbance readings were started. The reference cell normally contained the same solution as the sample cell, except for the substance being measured. For the halogenation of the arylacetylenes. the appearance of the product phenylhaloacetylenes was followed; for 3-methylbutyn-3-01 (MB). the disappearance of sodium hypohalite was followed. Detailed spectral data are found in the thesis's and are summarized in Table I. We found, for example. that Beer's law (eq 6) held for

Table I. Ultraviolet Spectral Data of Arylacetylenes, 4-YCaH4C-H. and Their Halogenation Products, 4-YC6H4C=CX. in Methanol Y

H H H H CH3 CHa C Ha CI CI Br Br NO2 NO?

X

A,,,,.

H CI Br I H CI Br? H Br H Br H Br

nm

lo%, M-1 cm-'

244.5 250 253 256 249.5 253.5 256 252 260 254.5 262 292 300b

This value is not significantly changed in water.

OD

=

log ZojZ

=

ebc

0 . 91a

0.98 1.15 1.59 1.72 1.95 2.07 2.30 1.92 2.11

Reference 13.

(6)

phenylchloroacetylene, phenylbromoacetylene, and sodium hypochlorite from 0 to 10-4,lO-3, and 10-2 M , respectively. All of the kinetic runs were carried out under pseudo-first-order conditions. That is, for the conditions [OHIO > [C~O-IO>> [ArC=CH], the pseudo-first-order expression

rate

=

k+[ArC=CH]

(7)

(20) I. M. Kolthoff and E. B. Sandell, "Textbook of Quantitative Inorganic Analysis," 3rd ed, London, Macmillan, New York, N. Y . , 1952, chapter XXXIX.

Lii, Miller 1 1-Haloalk-ynesfrom Acetylenes and Sodium Hypohalites

1604 were made up (25.0’) to t a t which the run was carried out. Because we ended up with a third-order rate law, the adjustment is given by kcm

=

kobsd(&/dt)

(9)

in which dwas taken as the density of water. The activation parameters were determined from a least-squares treatment of log k / T DS. l/T according to the standard relation23

k

2>j ‘I”’/EZer&?r ,

qL1

Figure 1. The ultraviolet spectra of phenylacetylene (A) and Its halogenation products, phenylchloroacetylene (B) and phenylbromoacetylene (C) In methanol. applied. obtain

By combining eq 6 and the integrated form of eq 7, we

2.303 log (OD, - OD)

=

-k+t

+B

(8)

in which OD is the optical density of the solution, tis the time, and B is a constant. An analogous procedure was followed when hypohalite was monitored. Note that unless OD, depends on it, the initial concentration of the substance being monitored does not need to be known precisely, The kinetic runs proceeded for ca. 4 half-lives. When the disappearance of a reactant was followed, the optical density at “infinite” time was taken after at least 10 half-lives. Since this OD, was usually small (-0.024.06), this leads to a difference of 3-8 in the rate constants as calculated for the observed OD, and the expected OD, = 0. When the appearance of the haloalkynes was followed, the observed OD, was found to be somewhat low, as compared with the value expected from the known e value. The discrepancy in (OD),, which presumably arises because of the lability of the haloalkynes. leads to -10% difference in the estimates of the rate constants for a single run. Two compounds of interest to us were not suitable for study by our techniques. The bromination of p-nitrophenylacetylene appeared to have the right stoichiometry according to eq 1, but A,, of reactant and product overlapped; since the solubilities of these M ) , it was difficult to devise an altercompounds were low nate to the uv method. With respect top-methoxyphenylacetylene, bromine addition to the triple bondz1or perhaps attack on the ringsZ appear to compete with process 1, so that this system too was not studied kinetically. For each system, at least five runs were made at different concentrations of reagents. The mean rate constant was then corrected for volume changes from the temperature at which the solutions (21) (a) J . A . Pincock and I 2[ClZ],, an aqueous solution of phenylacetylene (1.74 X 10-4 M ) , sodium hydroxide (0.65 M ) , and sodium hypochlorite (0.04 M ) at 25” developed a n OD at A 250 nm which indicated essentially quantitative conversion (-95 %) to phenylchloroacetylene. Likewise, bromination runs of the arylacetylenes yielded solutions whose uv spectra were identical in shape with those of the expected arylbromoacetylenes; the calculated yields usually exceeded 90 %. At least some of the deviation from the theoretical yield may be ascribed to the side reactions of the following paragraphs. At comparable hydroxide and chlorine concentrations, [OH-Io ‘u [Cl2l0,the yield of phenylchloroacetylene was reduced. A solution of phenylacetylene (1.5 X M ) , sodium hydroxide (0.21 M ) , and chlorine (0.1 M ) at 46.1 gave a maximum of -25% phenylchloroacetylene, which then decreased slowly with time, according to uv analysis at 250 nm. Judging by the uv spectrum of authentic 226.5 nm), this compound appeared in the reacbenzoic acid ,,A,( tion solution after 40 hr. As a check, the above reaction was scaled up to 1.5 1. and the product mixture worked up. The presence of benzoic acid was established by tlc. Chlorine solutions that are deficient in base, [OH-Io < [Cl&, d o M) not give process 1. Solutions of phenylacetylene (-8 X and chlorine (-0.05 M ) ,with and without sodium hydroxide (-0.05 M ) , present at 46.1 ’, yielded a product solution whose uv spectrum 257.5 nm). A prepararesembled that of e,P-dichlorostyrene (A,, tive scale reaction of phenylacetylene (2 ml), chlorine (2 ml), and 0.02 M sodium hydroxide (250 ml) as a heterogeneous mixture was carried out at -25” on a shaking device. The product was worked up and identified by uv and nmr spectra as cis- and traiis-a&dichlorostyrene.

,L

240I

“,b

= (kT/h)eAS*/Re

1010.

Journal of the American Chemical Society

Results and Discussion Stoichiometry. The preparation of some 1-haloalkynes from terminal alkynes according to eq 1 is well documented. Phenylbromoacetylene (80 yield) and 11 of its ring-substitution products (unspecified yields) were obtained routinely in this way.13 Under conditions similar to those of our kinetic runs, we established that phenylchloroacetylene was produced in essentially quantitative yield, and four arylbromoacetylenes and phenyliodoacetylene were produced in yields 3 90 %, as determined by uv spectral analysis. The halogenation process 1 is favored only under specified conditions. Depending on the pH, one might have to contend with one or more of the following forms of halogen, e . g . , X,, X3-, HOX, and OX- (and perhaps others), each of which has its characteristic reactivity. In strongly basic aqueous solutions (pH > 1l), it is possible to have hypohalite ion as the dominant halogenating species. For here the pertinent equilibrium is 20H-

+ XZ+XO- + X- + HzO

(11)

and the equilibrium constants at 25” are K ‘v lo1:, lo8, and 30 for Xz = Clz, Br,, and I?, respe~tively.’~One cannot suppress any one of these species completely, and if its reactivity is sufficiently high, it will compete with hypohalite for the alkyne, Such competition was probably present in the p-methoxyphenylacetylenehypobromite system, where electrophilic attacks on the (23) A. A . Frost and R. G. Pearson, “Kinetics and Mechanism,” 2nd ed, Wiley, Ne% York, N. Y . , 1961, Chapter 5 .

95:5 1 March 7, 1973

1605

double bond or ring are expected to be fast.21f22v24 Now consider our findings for phenylacetylene in reaction 12 in three situations: a, [OH-Io > 2[C12]o;b,

I

I

I

I

I

I

I

b

(C6H,C(OH)&HCl)

+

i

CtjHsCOOH CeHsCE&H

1

+ OH- + Clz

ic+

C6HsCCkCHCl

a

CeHjCECCl CsH~CCkCC12

(12)

[OH-], 'v 2[Cl2l0;c, [OH-]@< 2[Cl2l0. Condition a was used in our kinetic studies. Condition b led to reduced yields of phenylchloroacetylene at short reaction times and to benzoic acid over a long period. Condition c led to a,P-dichlorostyrene. It is clear that solutions deficient in base give characteristic addition reactions of chlorine or hypochlorous acid to an unsaturated comp ~ u n d . ~This ' ~ ~is~ in contrast to the behavior of many weak carbon acids, e.g., ketones, in which a change in pI1 merely alters the mechanism, not the process. In this work, a drop in p H can be catastrophic. Kinetics and Base Mechanism. The arylacetylenes and arylhaloacetylenes are sparingly soluble (< M) in water. Fortunately the uv spectra (Table I, Figure 1) of reactant and product of eq 1 did not overlap appreciably so that this method of following the reaction could be used. In the case of p-nitrophenylacetylene, of it and its product overlapped however, the , , ,X strongly; in addition, its low solubility rendered other methods impracticable. Unlike ordinary alkylacetylenes, methylbutynol (MB) had fair solubility in water. We could therefore follow its reaction by monitoring hypohalite ion by uv. The kinetics of chlorination of this compound, which had some unusual features, will be discussed separately in the Appendix. Since we were not committed to any particular rate law, we sought the kinetic order in each reagent. In the uv method, it is convenient to use pseudo-first-order conditions. Our concentration ranges were, therefore, [OH-Io > 2[X2l0> 100IArC=CH]o. However, we did change the initial concentrations in order to establish the kinetic order in each of the participants. Typical runs (Figure 2, Table 11) establish reaction 1 as first order in acetylene. The variation in acetylene concentration leaves the pseudo-first-order rate constant unchanged, as it should, and incidentally indicates the precision in k,. By keeping the [NaOH]@constant and varying [OX-], we found the order in hypohalite to be unity (Figure 3, Table 11). Then by keeping hypohalite constant and varying the sodium hydroxide, we found the kinetic order in NaOH to be unity (Figure 4). Rate law 2 applies to our systems. (If one postulates [X,] or [HOX] as the halogenating agent in place of [OX-], one will end up inevitably with the wrong rate law.) Our rate data are given in Tables I11 and 1V. As a result of these kinetic studies, we shall take as our working hypothesis that eq 3 applies; a preequilibrium step, for which K, = k,/k-, and the base is hydroxide ion, is followed by a rate-determining halogen transfer from hypohalite ion. The observed rate con(24) (a) S . Goldschmidt, R. Endres, and R. Dirsch, Chem. Ber., 58, 572 (1925); (b) R. 0. Norris, R. R. Vogt, and G. F. Hennion, J . Amer. Chem. SOC.,61, 1460 (1939); (c) J. Koenig and V. Wolf, Tetruhedron Lett., 4629 (1970).

Figure 2. Typical pseudo-first-order plots for the reaction of 4-YC&14C=CH with bromine in aqueous sodium hydroxide: 0, Y = H, 27.5"; A, Y CH3, 35.3"; 0,Y = C1, 25.3"; q ,Y = Br, 25.3 '. Only ca. 2 of theca. 4 half-lives followed are shown .

Table 11. A Test of the Third-Order Dependence of the Reaction of Sodium Hypochlorite with Phenylacetylene at -25' 105. [C6HjC=CH], [NaOH],

103. [OCl-1,

M

M

M

5

18 25 31 32 19 13 12 18

1.77 1.77 1.77 1.77 1.77 1.73 1.77 1.79 1.80

47 47 47 47 47 46 24.5 16.5 9.2

15 16.5 17 16 17 16 16

0.230 0.069 0.640 0.309 0.680 0.890 0.518

46.7 46.5 39.7 19.9 22.7 23.2 9.9

104k+, min-l

105k, M-2 min-1

122 114 120 120 122 121 60 42 23

147 137 144 144 147 152 138 143 139 Meana 143 f 4 6.0 55.9 1.9 59.0 14.9 58.5 3.2 51.7 9.0 58.6 12.4 60.2 2.89 56.7 Meanb 57 Zk 2

At 25.5". Since the ionic strengths in these solutions were high, these k's are larger than those below and in Table 111. See text for a discussion of medium effects. At 25.3".

stant is composite k o b s d = kbKa/[H2O] (13) The kinetic implications are easily derived. From the pK's of water, 15.7, and phenylacetylene, 21,25aand k, /v 3 X I O 3 M-I sec-I (from exchange of labeled p h e n y l a ~ e t y l e n e ) ,we ~ ~ ~can make the following estimates: K, N 10-6.3,k-a N 6 X IO8 M-I sec-', and kb = lo8 M-I sec-I for the bromination of phenylacetylene at 25". In several succeeding sections we shall see how useful this over-all mechanism is. Structure and Reactivity. The ratio of the third-order rate constants at 25.3" for the reaction of hypohalites with phenylacetylene is IO-, : k[Bro-): k(clo-, = 3.4 :1.2: 1.2 X This gives the reactivity of hypohalite ions with phenylethynyl anion as IO- > BrO- >> C10-. Although maxima and minima in the properties of

(25) (a) A . Streitwieser, Jr., and D. M. E. Reuben, J . Amer. Chem. Soc., 93, 1794 (1971); (b) E. A. Halevi and F. A . Long, ibid., 83, 2809 (1961).

Lii, Miller

1 1 -Haloalkynes from Acetylenes and Sodium Hypohalites

1606 Table III. Rate Data for the Halogenation of Substituted Phenylacetylenes ( 4 - Y C e H 4 B C H ) or 3-Methylbut-1-yn-3-01 (MB) with Sodium Hypohalite (NaOX) in Aqueous Sodium Hydroxide

X (concn, M )

Y (lo6 concn, M )

c1

H (10-17)

0.07-0.89

Hb (15-21) H (1.4-9.5)

0 . 3 - 0 , 63b

(0.026-0.076) Br ((38-525) X Br ((8-28) x 10-4) Br ((98-405) x 10-5)

Hb (7-8) CH3 (4-9.1)

0.026-0. 76b

Br ((4-8)

c1

0.04-0.047

(3.1-6.9)

(0.01-0.047)

c1

I ((5-28) Br ((8-30)

c1

0.034-0.175

25.3 36.0 46.1 46.1

7 7 8 4

5.8 16.0 27.5 36.0 27.5

4 8 14 8 5 4 5 5

Br (3.2-4.3)

0.04-0. 045

10-4)

x

H (7-11)

0,043-0.098

10-4)

x

MB

0.036-0.14

25.3

6

10-4)

(11-60) X lo-' MB (12-31) X lo-'

0.29-1.15

25.3

3

(0.012-0.01 5 )

k,a M-%sec-l

runs

x

10-4)

0.04-0.045

No. of

15.3 25.0 35.3 15.3 25.3 35.3 15.3 25.3 35.3 25.3

x

Br ((7-43)

Temp, "C f 0.1"

[NaOH], M

(95 (211 (431 (428

32 3) x 10-6 32 5) x 10-5 -f 17) X

f 10) x

3.75 f 0.23 6.89 0.21 11.34 i 0 . 4 2 17.36 f 0.48 10.57 f 0 . 2 4

*

4.40 7.31 12.74 9.63 15.11 24.25 10.02 15.67 25.26 32.12

5

4 6 6 5 5 6

f 0.03 f 0.18 f 0.3 f 0.35 f 0.37 f 0.45 f 0.2 i 0.5 f 0.55 f 0.70

2.02 i 0.01 (2.12

+ 0.11) x

10-4

a The average deviation, which is < 5 % , is indicated. These figures have not been rounded off so that our activation parameters may be reproduced. This run had C&jC=CD and NaOD in Dz0. The isotope rate effect was for chlorination, k c ~ ~ / k c n== 1.01 f 0.04; for bromination, kcaH/kcnD = 1.07 i 0.04.

Table IV. Activation Parameters at 25.3 for the Reaction of Arylacetylenes, 4-YCsH4CkCH,with Halogens (Xd in Aclueous Sodium Hydroxides

Y

X

k , M-? sec-'

H H CH,

c1

95 x 10-5 10.4 7.5 15.1 15.7

c1 Br a

Br Br Br Br

Log A , M-1 sec-l 6.9 7.3 7.7 7.1 7.2

+ 0.4

13.7 8.6 9.3 8.1 8.1

-ir 0 . 3

i 0.3

f 0.3 & 0.3

AH*, kcal/mol

A E,, kcal/mol f 0.4 i 0.2 i 0.3 f 0.3 f 0.3

13.0 8.0 8.7 7.5 7.6

-AST, eu

f 0.4

27 27 26 28 28

f 0.2 f 0.2 f 0.3 i 0.3

The probable errors of the activation parameters are included. I

Figure 3. The reaction of 4-methvl~henvlacetvlenewith sodium hypobromite in water at a fixed [NaOH], = 0.045 M: 0,15.3"; A, 25.0"; 0, 35.3".

halogen species are k n o ~ n , ' the ~ ' ~preceding ~ is the order of bimolecular disproportionation rate constants (to halide and halate ions)16-18 and of polarizabilities, ( 2 6 ) M. Eigen and K . Kustin, J . Amer. Chem. Soc., 84, 1355 (1962).

Journal of the American Chemical Society / 95:s

I

I

,

I

,

I

Figure 4. The reaction of phenylacetylene (ca. loe4 M ) with sodium hypochlorite in water at a fixed [NaOClIo = 0.047 M: 0, 25.3"; A, 36.0"; 3,46.1°.

etc., but the reverse of the usual order as oxidants based on standard potentials.1g Probable trends in the reactivity of carbanions with free halogens and with trihalides are If > Br2 > C1, and Is- > Br3- > Cl3-;*' it (27) (a) R. P. Bell and P.Engel, J . Chem. SOC.,247 (1957); (b) R . P. Bell and D. C. Vogelsong, ibid., 243 (1958).

1 March 7, 1973

1607

does appear that rates toward enolate anion^^,^^ are in the order Xz > X3- > OX- or that hypochlorite is the least reactive of this group. To compare aliphatic with arylacetylenes, we also investigated the base-catalyzed halogenation of MB. The ratio of the third-order rate constants at 25.3” of phenylacetylene t o MB for chlorination and bromina‘v 5. This is a tion is k(ArC=CH):k(RC=CH) modest effect. Recall that the overall rate constant is composite (eq 13). Any inductive and resonance effects which favor the aryl group in step 1 (K,)would retard the halogenating step (kb). Judging from our results, the effects on the two steps are roughly compensatory. Turning to the activation parameters (Table IV), one must again take into account the composite nature of kobsd(eq 13). We shall show that the interesting asAS*obsd

=

AS,

AH*obsd

=

AH,

(14)

AS*b

(15)

AHfb

pects of the observed parameters, except possibly for a constant term, may be identified with those obtained for the second-order process, the rate-determining step kb. The equilibrium process in eq 3 may be regarded as deriving from the two dissociations ArCECH H20

+ H+ + H+

ArCEC-

(16)

OH-

(17)

Since AS17 = -19.22s and ASl6 is probably .u-20 also,28we guess that AS, = ASl6 - As17 N 0 eu. On this basis, it would be expected that AS*,b,d should be characteristic of second-order, negative ion-negative ion associations. Judging from both similar and different charge types,28e . g . , C10C10- with AS* = - 19.6, HS06Br- with AS* = -24.8, BrCH2COOS2032-with AS* = -28.3 eu, and A r C E C T OHwith AS* -42,29 our values of A S * in Table IV are quite reasonable. The chlorination and bromination of phenylacetylene have about the same AS*, but differ by -5 kcal/ mol in AH*. This could be taken to mean that the mechanisms are identical but that the reactivities of hypochlorite and hypobromite are very different. For the bromination of the four phenylacetylenes the variations in AH* and AS* are quite small. Nevertheless, a good fit to the Hammett equation was obtained, yielding p values of 0.718,0.766, and 0.825 at 35.3, 25.3, and 15.3O, respectively. The temperature dependence of p is given by eq 18,10,30 in which the coefficients were ob-

+

+

-

p =

+

-(0.823

0.095)

+ (475

+

28)/T

(18) tained by a least-squares fit.30 Furthermore, since log k varies either as u or as 1/T, it can be given rigorously by a concise four-parameter equation. 30 &

&

+

1.012)~- (1847 i 52)/T (481 315)o/T 7.21 + 0.23 (19) Equation 19 was obtained by a computer fit of the data by a program which also gives the probable errors in the log k

=

-(0.829

f

*

+

(28) Reference 23, Chapter 7. (29) C. Eaborn, G. A. Skinner, and D. R. M. Walton, J . Chem. SOC. B, 922 (1966). (30) (a) S. I. Miller, J . Amer. Chem. Soc., 81, 101 (1959); (b) G. S. Krishnamurthy and S. I. Miller, ibid., 83, 3961 (1961).

coefficient^.^^ Although the fit of 13 k’s may seem to be too accommodating, the differences between observed and calculated k’s (not log k) are
1. Isotope Effect. Rate constants for the deuterioxidecatalyzed chlorination and bromination of phenylacetylene in deuterium oxide were determined by methods identical with those used in water solutions. The results are included in Table 111. (36) J . R. Jones and S. C. S. Rao, Trans. Faraday Soc., 63, 120 (1967).

Journal of’the American Chemical Society

95:s

+ +

+

(37) R. L. Schowen, Progr. Phys. O r g . Chem., 9,275 (1972). (38) E. Hogfeldt and J . Bigeleisen, J . Amer. Chem. Soc., 82, 15 (1960). (39) R . D. Guthrie, A. T.Young, and G. W. Pendygraft, ibid., 93, 4947 (1971); (b) G. Levin, C. Sutphen, and M. Szwarc, ibid., 94, 2652 (1972); (c) D. H. Hunter and S. K. Sim, Can. J . Chem., 50, 678 (1972). (40) C. G. Swain, D. A. Kuhn, and R. L. Schowen, J . Amer. Chem. Scc., 87, 1553 (1965); C. G. Swain, A . D. Ketley, and R. F. W. Bader, ibid., 81, 2353 (1959); P. Ballinger and F. A. Long, ibid., 81, 2347 (1959).

March 7, 1973

1609

anol has kH/kD = 0.91.41 Note that these systems resemble ours in that there is a preequilibrium; they differ in that the rate-determining step is unimolecular rather than bimolecular, and that the activated complex is singly rather than doubly charged. On this basis, it would seem that our kH/kD N 1.0 is the result of fortuitous cancellation rather than the absence of isotope rate effects. Unfortunately, we have no good models and/or reliable fractionation factors37 to estimate all of the individual contributions. This means that the choice among mechanistic alternatives, which we shall make later for step b of eq 13, cannot be guided or limited by the measured isotope effect. Halogenation Mechanism. Since the mechanism of the first step of process 3 seems to be firmly established, we turn to the subsequent steps which appear to be initiated by rate-determining attack of hypohalite ion on phenylacetylide ion. Several alternative paths to the final products are available, e.g., eq 24-26. In these,

+ XO- a A r M X + O*b + HzO e 20H(24) ArC=C- + XO- + H 2 0e ArCSCX + 20HArCEC- + XO- e A r C = C X 0 2 h ArC=CXO2- + H 2 0a A r C = C X O H - + OH(25) ArCZCXOH- +ArC-CX + OHd ArC=CXOz- + H 2 0e ArC=CX + 20H-

species we postulate. At this stage, we can find no compelling reasons to reject any of the species or steps involved either in path 25a-25b-25c, or in 26a-26b25c. The doubly charged transition state leading to ArC=CX02- in 25a deserves comment. Although Rappe considered and dismissed a negative ion-negative ion transition state in the reactions of enolate ions with h ~ p o h a l i t e such , ~ ~ a species seems to be required by the observed k i n e t i ~ s . ~There ,~ is no difficulty in finding precedents for reactions of similarly charged ions, e.g., Br3- e n o l a t e ~XO, ~ ~ XO-,16,17S2032CICHzCOO-, ** BrOCz042-,4 6 etc. A second factor which bears on the existence of ArC=CX02- may be the so-called a effect.47 Hypohalite is a reactive species, a “superphile,” by virtue of the lone pairs of electrons on the atom adjacent to the attacking center.47 In the ground state, the filled p atomic orbitals are thought to form two molecular orbitals, e.g.

+

Ar-

02-

-

C

of ao 0

a

C

a

OXArC=CArC=C-

+ H20 eX(OH)z-

+ X(OH)2- e AI) r C = C X O H - + OH+ X(0H)z- e ArC=CX + 20H-

(26)

C

+

+

+

-WHoMo

+

ao

oc1-

Thus, the HOMO is relatively high and XO- relatively rea~tive.4~Such a effects have precedent in the reactions of hydroperoxide, hypohalite, etc., with 4-nitrophenylacetate. 47 Thirdly, both reacting ions, ArCGCand XO- have no K stabilization energy on the simple Huckel MO model, but the intermediate, A r G C X O z-, formed from them has twice the Huckel stabilization energy of butadiene (see Scheme I). In this diaScheme I

we make all steps reversible both because we shall consider these specifically and because we wish to point to the generality of such forward and reverse processes. In evaluating these mechanisms we have tentatively rejected steps which appear to be implausible in at least one direction. Because of its high energy, we considin (24a) i m p r ~ b a b l e . Since ~~ ered the formation of 02eq 24c and the reverse of eq 24c, 25d, and 26c are termolecular, we regard these as far less probable than alternative stepwise processes. We are left with the mechanistic sequences (25a,b) or (26a,b), followed by (25c). Both of these involve doubly coordinated positive halogen ions, e.g., X(OH),-, ArC=CXOH-, and ArC=CX02-. The first of these is obviously related to trihalide or analogous ions such as X20H- and XYOH- which have in fact been proposed as intermediates in reactions of hypoh a l i t e ~ . ’ Acetylide ~ ~ ~ ~ ~ (like ~ ~ cyanide) may be regarded as a pseudo halide and, therefore, is included in the group. One can also point to stable haloacetylenebase complexes, R C c C - X . . .B, e.g., phenyliodoacetylene with aniline or m ~ r p h o l i n eas , ~models ~ for the (41) R. A. More O’Ferrall and S. Slae, J . Chem. SOC.B, 260 (1970). (42) C. S. G . Phillips and R . J . P. Williams, “Inorganic Chemistry,” Vol. 11, Oxford University Press, London, 1966, p 502, give A H f ” ( O 2 - )= 217 kcal/mol for reaction 24b in the vapor phase; AH ‘v -224 kcal/mol for eq 24b. (43) M. Anbar, S . Guttmann, and R . Rein, J . Amer. Chem. Soc., 81, 1816 (1959).

ArC

--C-

xo-

(ArC=CXO)?-

gram, we show only the rz orbitals of A r C r C - , and r2’for XO-. We assume that the ir2 and r2’ energy levels are very close; this would not change appreciably for a more exact treatment. Using a butadiene model for our intermediate, we find 1.24 /3 stabilization energy for r2 orbitals. This figure should be doubled, if all of the T electrons are considered. Therefore, the transition state (ArCGCXO ”) may not be unreasonable. (44) J. U . Nef, Justus Liebigs Ann. Chem., 308, 264 (1899); P. L. Southwick and J. R. Kirchner, J. Org. Chem., 27, 3305 (1962); R . H . Baughman, ibid., 29, 964 (1964). (45) C. Rappe, Acta Chem. Scand., 22, 219 (1968). (46) K . C. Grover and R. C. Mehrotra, 2.Phys. Chem. (Frankfurt am Main), 69, 308 (1970). (47) G. Klopman, K . Tsuda, J. B. Louis, and R. E. Davis, Tetrahedron, 26,4549 (1970); J. D. Aubort and R . F. Hudson, Chem. Commun., 937 (1970).

Lii, Miller J 1-HaIoaIkynes from Acetylenes and Sodium Hypohalites

1610

Kinetic Order of Halogenations. It is necessary to emphasize that second-order kinetics for the halogenation of weak carbon acids is generally accepted and is often valid. Why then do we have a third-order rate law? Our answer in general terms is that these two rate laws reflect numerous mechanistic variations in halogenation mechanisms as one passes from very low to very high pH values. As the pH changes, the number, kind, and reactivity of halogenating species change. Of course, the same applies to the weak carbon acid species. Unless this is taken into account, qualitative observations and even rate laws may seem confusing; e.g., phenylbromoacetylene can be produced from both acidicz1and basic solutions containing phenylacetylene and bromine. If we confine ourselves to the basic range, then the second-order kinetics can usually be associated with the fast reactions of the carbon acid anion with X3-, X,, or XOH. Certain factors, however, may contribute to the incursion of the third-order rate law (eq 2). First, the weaker the carbon acid, the closer to the encounter rate will be the reaction of the carbanion (C-) with water (k-,). Therefore, competition from the halogenating step is less effective. Second, at high pH, [XO-] >> [X,], and the fast reaction of X, hith C- is negligible ; thus the conditions for second-order kinetics are unfavorable. Third, chlorine is "slow," judging from the large gap in our hypobromite and hypochlorite rates; it is conceivable that in reactions with iodine, bromine, or chlorine at high pH, a given carbon acid may follow either a second- or third-order rate law, or perhaps both. An exact balance between second- and third-order kinetics, or "equipartition" as we have called it,6 occurs when the two terms in the denominator of eq 4 are equal, that is [Ox-lec = [H?O]k-a/kb (27) Alternatively, the equipartition condition (ec) applies to that concentration of hypohalite ion above which second-order, and below which third-order, kinetics dominate. In the case of phenylacetylene at pH >11 and 25", it can be shown that [OCl-Iec = 3 X lo6, [OBr-I,, = 2 X lo2, and [01-I,, = 9 X 10 M e 6 1-Dehalogenation of Positive Halogen. If process 1 is reversible and if the kinetics (eq 2) and the mechanisms we proposed are correct, then the simplest kinetics of the reverse process, e.g. C6H5C=CBr

+ 20H- *C6H5C=CH + OBr- + OH-

Table VI. Rate Data for the Reaction of Phenylbromoacetylene with Arsenious Oxide in Aqueous Sodium Hydroxide at 26.0" 1O3[AszO3], [NaOH], M M 3.13 3.13 1.56 1.56 3.13 0.78 6.25

(48) J. I. Dickstein and S . I. Miller, J . O r g . Chem., 37, 2168 (1972).

0.26 0.12 0 26 0.064 0.062 0.26 0.052

104k+, sec-l

k, M - sec-

5.98 1.29 2.87 1.79 0.34 1.47 0.47 Mean

2.83 2.86 2.72 2.80 2.83 2.78 2.76 2.80 f 0.04

indicate a curious rate law in the concentration range examined, namely, first order with respect to arsenious oxide, first order with respect to phenylbromoacetylene, and second order with respect to hydroxide ion. - d[ArC=CBr]/dt

=

k+[ArC=CBr] = k[ArC=CBr][Asz03][0H-]

*

(29)

When the concentration of hydroxide ion was larger than 0.26 M , the product differed from that of pure phenylacetylene, and we did not investigate the reaction further. Since it has been reported that a relatively large amount of the arsenious oxide must be in alkaline solution in the form A S O , - , ~ one ~ of several possible mechanisms might be

+

(28)

would be first order in haloalkyne and second order in hydroxide ion. For this reason, we thought that a study of 1-dehalogenation of haloalkynes could be instructive. Dehalogenations such as process 28 are well known in the chemistry of haloalkynes. 3, 13,l 5 ,48 Indeed, abstractions of positive halogen by methoxide, l 5 tributylphosphine, 48 or ethanethi~late'~are first order in each reagent. What is special about process 28 is the requirement for third-order kinetics. An analogy for this unusual situation is found in the deiodination of iodomalonate esters, in which the contribution of the term c ' = k[I-lZ [ICH(COOR)2]to the total rate is important.27" To reverse process 1, we sought a scavenger which

Journal o j the American Chemical Society / 95:s

would react with hypohalite as it formed and would not otherwise alter the system. Accordingly, we examined phenylbromoacetylene and several scavengers 253.5 nm and the looking for the disappearance of A, 244.5 of phenylacetylene at 26". appearance of A, Suffice it to say that phenylbromoacetylene (10-4M ) either did not react with sodium hydroxide (ca. 2 M ) in ca. 2-15 hr at 26" in the presence of reductant (